The Effect of Short-Time Fourier Transform Parameters Choice on ENF-based Forensic Analysis of Audio

Öz

Due to the rapidly developing computer technology, digital audio, image, and video can be modified with various techniques. These modifications may be on the media content or the metadata. In this context, forensic analysis of digital media is of great importance. Electrical network frequency (ENF) based forensic analysis is a significant tool that can be used for checking file integrity and detecting time-of-recording. One of the most preferred methods for ENF signal estimation is the Short-Time Fourier Transform (STFT) based approach. The choice of STFT window size and STFT hop size can directly affect the accuracy of ENF signal to be estimated from media, and hence the performance of the ENF-related forensic applications. This work investigates how the STFT parameters choice affects the performance in time-of-recording verification. A comparative analysis is made for various STFT window sizes and STFT hope sizes by experimenting with different audio lengths.

Anahtar Kelimeler

• ENF
• Electric Network Frequency
• Audio and Video Forensics
• Time-of-Recording
• Time-Stamping

Ses Dosyalarının ENF Tabanlı Adli Analizine Kısa Zamanlı Fourier Dönüşümü Parametre Seçimlerinin Etkisi

Öz

Hızla gelişmekte olan bilgisayar teknolojisi sayesinde çeşitli tekniklerle dijital ses, görüntü ve video dosyaları üzerinde modifikasyonlar yapılabilmektedir. Bu modifikasyonlar doğrudan dosya içeriğinde olabileceği gibi bazen de meta data üzerinde olabilmektedir. Bu bağlamda dijital medyaların adli analizi büyük önem arz etmektedir. Elektrik şebeke frekansı (ENF) tabanlı adli analiz yaklaşımı dosya bütünlük kontrolünde ve dosyaların kayıt zamanı tespitinde kullanılabilen önemli bir araçtır. ENF sinyali kestiriminde en çok tercih edilen yöntemlerden biri, kısa zamanlı Fourier dönüşümü (Short-Time Fourier Transform - STFT) temelli yaklaşımdır. STFT yönteminde, pencere boyutu ve kaydırma miktarı parametrelerinin seçimi büyük öneme sahip olup, kestirimi yapılan ENF sinyali doğruluğunu, dolayısıyla da ENF tabanlı adli analiz uygulamalarının performansını doğrudan etkileyebilmektedir. Bu çalışmada, STFT parametreleri seçiminin, ENF tabanlı dosya kayıt zamanı doğrulamada performansa ne derece etki ettiği araştırılmıştır. Farklı uzunluktaki ses dosyaları, çeşitli STFT pencere boyutu ve STFT kaydırma miktarlarına göre ayrı ayrı test edilerek karşılaştırmalı bir analiz yapılmıştır.

Anahtar Kelimeler

• ENF
• Elektrik Şebeke Frekansı
• Ses ve Videoların Adli Analizi
• Kayıt Zamanı
• Zaman Damgası

Kaynakça

1. M.H. Bollen and I.Y. Gu “Signal Processing of Power Quality Disturbances,” John Wiley and Sons, 2006.
2. C. Grigoras “Digital audio recording analysis–the electric network frequency criterion”, International Journal Speech Language Law, vol. 12, no. 1, pp. 63–76, 2005.
3. A. Cooper, “The electric network frequency (ENF) as an aid to authenticating forensic digital audio recordings–an automated approach”, Audio Engineering Society Conference, pp. 1–10, 2008.
4. E.B. Brixen “Techniques for the authentication of digital audio recordings”, Audio Engineering Society Conference, vol. 122, pp. 1–8, 2007.
5. J. Chai, F. Liu, Z. Yuan, R. Conners, and Y. Liu “Source of ENF in battery-powered digital recordings”, Audio Engineering Society Conference, vol. 135, pp. 1–7, 2013.
6. N. Fechner and M. Kirchner “The humming hum: Background noise as a carrier of ENF artifacts in mobile device audio recordings”, 8th International Conference IT Secure Incident Management IT Forensics, pp. 3–13, 2014.
7. S. Vatansever and A.E. Dirik “Forensic analysis of digital audio recordings based on acoustic mains hum”, 24th Signal Processing Communication Application Conference, pp. 1285–1288, 2016.
8. R. Garg, A.L. Varna, and M. Wu “Seeing ENF: Natural time stamp for digital video via optical sensing and signal processing”, 19th ACM International Conference Multimedia, pp. 23–32, 2011.

9. R. Garg, A.L. Varna, A. Hajj-Ahmad, and M. Wu “Seeing ENF: Powersignature-based timestamp for digital multimedia via optical sensing and signal processing”, IEEE Transactions on Information Forensics and Security, vol. 8, no. 9, pp. 1417–1432, 2013.
10. H. Su, A. Hajj-Ahmad, R. Garg, and M. Wu “Exploiting rolling shutter for ENF signal extraction from video”, IEEE International Conference Image Processing, pp. 5367–5371, 2014.
11. M. Wu, A. Hajj-Ahmad, and H. Su “Techniques to extract ENF signals from video image sequences exploiting the rolling shutter mechanism; and a new video synchronization approach by matching the ENF signals extracted from soundtracks and image sequences”, U.S. Patent US9916857B2, 2015.
12. S. Vatansever and A.E. Dirik “Videoların ENF tabanlı adli kanıt analizine ışık kaynağı etkisi”, Journal of Uludag University Faculty of Engineering, vol. 22, pp. 53–64, 2017.
13. S. Vatansever, A.E. Dirik, and N. Memon “Detecting the presence of enf signal in digital videos: A superpixel-based approach”, IEEE Signal Processing Letters, vol. 24, no. 10, pp. 1463–1467, 2017.
14. S. Vatansever, A.E. Dirik, and N. Memon “Analysis of rolling shutter effect on enf based video forensics”, IEEE Transactions on Information Forensics and Security, vol. 14, no. 9, pp. 2262–2275, 2019.
15. S. Vatansever “Modern techniques in forensic analysis of multimedia signals”, Ph.D. Dissertation, Deptartment of Electronics Engineering, Bursa Uludağ University, Bursa, 2019.
16. J. Choi and C.-W Wong “ENF signal extraction for rolling-shutter videos using periodic zero-padding”, IEEE International Conference on Acoustics, Speech, and Signal Processing, pp. 2667–2671, 2019.
17. S. Fernández-Menduiña and F. Pérez-González “Temporal localization of non-static digital videos using the electrical network frequency”, IEEE Signal Processing Letters, vol. 27, pp. 745–749, 2020.
18. P. Ferrara, G. Draper-Gil, I. Sanchez, H. Junklewitz, and L. Beslay “Modelling gop structure effects on ENF-based video forensics”, International Conference on Digital Forensics and Cyber Crime, pp. 121–138, 2020.
19. D. Nagothu, Y. Chen, A. Aved, and E. Blasch “Authenticating video feeds using electric network frequency estimation at the edge”, EAI Endorsed Transactions on Security and Safety, vol. 7, no. 24, pp. 168648, 2021.
20. D. Bykhovsky and A. Cohen “Electrical network frequency (ENF) maximum-likelihood estimation via a multitone harmonic model”, IEEE Transactions on Information Forensics and Security, vol. 8, no. 5, pp. 744–753, 2013.
21. S. Vatansever, A.E. Dirik, and N. Memon “ENF based robust media time-stamping”, IEEE Signal Processing Letters, vol. 29, pp. 1963–1967, 2022.
22. G. Hua and H. Zhang “ENF signal enhancement in audio recordings”, IEEE Transactions on Information Forensics and Security, vol. 15, pp. 1868–1878, 2020.
23. G. Hua, H. Liao, H. Zhang, D. Ye, and J. Ma “Robust ENF estimation based on harmonic enhancement and maximum weight clique”, IEEE Transactions on Information Forensics and Security, vol. 16, pp. 3874–3887, 2021.
24. C. Grigoras, A. Cooper, and M. Michalek “Best practice guidelines for ENF analysis in forensic authentication of digital evidence”, Forensic Speech Audio Analysis Working Groups, vol. 1, no. 1, pp. 1–10, 2009.
25. A. Hajj-Ahmad, R. Garg, and M. Wu “Instantaneous frequency estimation and localization for ENF signals”, Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, pp. 1–10, 2012.
26. J.G. Proakis and D.G. Manolakis “Digital Signal Processing: Principles, Algorithms, and Applications”, 4th Edition, Pearson, 2007.
27. A. Hajj-Ahmad, R. Garg and M. Wu “Spectrum Combining for ENF Signal Estimation”, IEEE Signal Processing Letters, vol. 20, no. 9, pp. 885-888, 2013.